On Wed, Aug 26, 2009 at 7:45 PM, Massimiliano Maini
<address@hidden> wrote:
>That is why getting a strong 0-ply is of the utmost importance.
>I am not sure what is the state of other bots.
We do agree on this, getting a strong 0ply NN is a must.
Actually that's why I was inquiring to know if the training
methodology of gnubg NN could be responsible of this kind of
behavior.
This whole odd/even effect thing seems to be much more visible
in gnubg than in other bots and it's the source of (possibly wrong)
beliefs like "gnubg 3ply cube is worse/better than 2ply cube"
or "2ply cube is worse/better than 3ply cube for closed out
positions" etc.
In the particular position (which is an extreme case):
GNU Backgammon Position ID: DwAAsN22AYgAAA
Match ID : MAHgAAAAAAAA
+-1--2--3--4--5--6-------7--8--9-10-11-12-+ O: gnubg
| O O O O O O | | | 0 points
| O O O O O O | | | On roll
| O | | |
| | | |
| | | |
| |BAR| |^ 7 point match (Cube: 1)
X | | | |
XX | X | | |
XX | X | | |
XX | X | X | |
XX | X O | X | O | 0 points
+24-23-22-21-20-19------18-17-16-15-14-13-+ X: MaX
O on roll:
Position ID: DwAAsN22AYgAAA
Match ID: MAHgAAAAAAAA
Evaluator: Crashed
Win W(g) W(bg) L(g) L(bg) Equity Cubeful
static: 76.8 0.0 0.0 0.1 0.0 +0.534 +0.785
1 ply: 88.0 0.0 0.0 0.1 0.0 +0.759 +0.923
2 ply: 79.7 0.0 0.0 0.1 0.0 +0.593 +0.999
3 ply: 87.7 0.0 0.0 0.0 0.0 +0.754 +1.000
4 ply: 82.0 0.0 0.0 0.0 0.0 +0.639 +1.000
BGBlitz has:
O on move: (Plies are starting with 1, so it's gnubg 0/1/2-ply)
1-ply Green 83.0 0.0 0.0
2-ply Green 81.9 0.0 0.0
3-ply Green 84.4 0.0 0.0
So BGBlitz too has some swinging between odd and even plies, but
the amplitude is 2-4%, not 12%. Snowie seems to behave like BGBlitz,
just 2-3% swing.
MaX.